Reducing response elimination strategies enhances the construct validity of figural matrices

Two studies are reported which explored the existence and effect of respondents' solution strategies in figural matrices problem solving on the construct representation and Gf saturation of the automatically generated figural matrices. The first study (N = 418) indicated that qualitative and quantitative differences in figural matrices problem solving exist and that the use of a response elimination strategy decreased the Gf-factor saturation of the figural matrices and affected the cognitive components involved in solving them. By contrast, the cognitive component processes involved in constructive matching strategies were in line with theoretical models and also yielded a higher Gf-factor saturation of the figural matrices. In addition, multi-group means and covariance structure analyses confirmed that response elimination constitutes a fallback strategy which is used if general cognitive control demands exceed the capacity of the problem solvers. Based on these findings the second study (N = 1890) examined systematic manipulations of the response alternatives as a means of preventing response elimination. The results indicated that using this theory-based method of constructing response alternatives can successfully enhance the construct validity of figural matrices tests.